MetaboT: An LLM-based Multi-Agent Framework for Interactive Analysis of Mass Spectrometry Metabolomics Knowledge
Abstract
Mass spectrometry metabolomics generates complex data that overwhelms traditional analysis approaches. MetaboT, a multi-agent Large Language Model (LLM)-based framework, converts natural-language questions into needed SPARQL queries, enabling effective knowledge graph navigation (https://holobiomicslab.github.io/MetaboT/). We validate MetaboT’s performance on a large plant dataset with 50 representative queries. MetaboT's modular design facilitates advanced data mining, biological interpretation, and the discovery of novel compounds without specialized programming expertise.
Related articles
Related articles are currently not available for this article.